GPT-6.1 Sol Ultrafast runs up to eight times faster and costs six times as much in the API

OpenAI began rolling out Ultrafast mode for GPT-6.1 Sol on October 8 in the API, Codex and ChatGPT Work. OpenAI says it delivers near-Astra intelligence at up to eight times the speed of standard Sol. In the API, set service_tier to ultrafast.

API pricing is $12 per million input tokens and $60 per million output tokens, six times the standard $2 and $10. Codex lead Tibo Sottiaux shipped it as day four of the 28-day Codex campaign, along with instant steering so the model reacts right away when you change direction mid-task. Some users on X complained that it favors the most expensive plan and API customers.